Satoshi Yoshiji, MD, PhD
satoshiyoshiji.bsky.social
Satoshi Yoshiji, MD, PhD
@satoshiyoshiji.bsky.social
Assistant Prof & PI at McGill University (@mcgill.ca) using human genetics & multi-omics for drug discovery | Endocrinologist | PhD at McGill & KyotoU | Visiting Scholar at the Broad Institute of MIT and Harvard & MGH
Twitter: https://x.com/satoshiyoshiji
Our commentary is out in Diabetes!
• MODY panels are undergoing refinement in the population-scale genetics era.
• We discuss how the Exeter team provides strong evidence for NEUROD1 and PDX1, but not for APPL1 or WFS1, using data from UK Biobank and gnomAD.
diabetesjournals.org/diabetes/art...
October 21, 2025 at 2:21 PM
Join us today at 4:40 pm at the Festival of Genomics & Biodata (@fogenomics.bsky.social) in Boston to explore how proteomics and genomics drive drug-target discovery!
June 24, 2025 at 3:39 PM
Honored to receive the Canadian Institutes of Health Research @CIHR_IRSC Catalyst Grant! We appreciate this valuable support for our research.
March 7, 2025 at 1:54 PM
Happy to be back at the Broad! Excited to reconnect with amazing colleagues and speak at the Diabetes Genetics Research Group😊
February 25, 2025 at 1:22 PM
If you’re in Philly attending US HUPO, join us for the morning session tomorrow at 7:15 AM on "Leveraging Publicly Available Large-Scale Proteo-Genomics Data to Identify Proteomic Drivers of Human Diseases: A Catalog of Proteome–Phenome-Wide Associations"

somalogic.com/us-hupo-2025/
February 23, 2025 at 10:24 PM
Interested in using proteomics to understand disease biology?
Join us at US HUPO 2025 in Philly for the Feb 24 SomaLogic breakfast session

I'll discuss our recent study in Nature Genetics on a protein mediating obesity’s effect on CAD & our MR atlas project.

somalogic.com/us-hupo-2025...
February 19, 2025 at 7:54 PM
📣 We are hiring!
Join Yoshiji Lab at McGill University Dept of Human Genetics in Montreal, Canada!
yoshiji-lab.org/joinus
✅Fully Funded Positions
• 2 MSc Students
• 1 PhD Student
• 1 Postdoc
✅Goal: Using human genetics and multi-omics for drug target discovery and precision medicine
January 26, 2025 at 2:00 PM
my cat reviewing my paper
January 24, 2025 at 2:59 PM
We’re hiring soon!
1–2 MSc positions, one PhD position (all fully funded), and a post-doc position will be available. Stay tuned and join us at McGill in the vibrant city of Montreal, Canada!

Lab webpage: yoshiji-lab.org
January 24, 2025 at 1:41 PM
Honored to work with these amazing people. Thank you so much!
10/10
January 24, 2025 at 1:41 PM
Overall, our study provides strong human genetic evidence supporting C-terminal COL6A3-derived endotrophin as a mediator of obesity’s effect on coronary artery disease and highlights it as a potential therapeutic target.
9/10
January 24, 2025 at 1:41 PM
Nice—but drug development takes time. In the meantime, can we do something? The answer is yes.
Using multivariable MR, we showed that reducing body fat (and for some proteins, increasing muscle) can lower protein mediator levels and cardiometabolic risk!
7/10
January 24, 2025 at 1:41 PM
We also found that the cis-pQTL of C-COL6A3 (rs11677932) overlaps with active enhancer regions in adipose tissue and the coronary artery. rs11677932 also disrupts DNA binding of a transcription factor, consistent with its association with ↓expression and plasma levels of COL6A3.
6/10
January 24, 2025 at 1:41 PM
In human tissue data from GTEx, COL6A3 was expressed in multiple tissues. We took a closer look at relevant tissues (adipose tissue and coronary artery) with single-cell RNA sequencing. COL6A3 was highly expressed in tissues involved in metabolic dysfunction and fibrosis.
5/10
January 24, 2025 at 1:41 PM
As orthogonal evidence, reassuringly, baseline C-terminal COL6A3 levels were associated with increased 10-year CAD risk in the UK Biobank.
4/10
January 24, 2025 at 1:41 PM
We closely examined COL6A3, which showed the highest OR for CAD.
Interestingly, domain-aware MR, enabled by two aptamers targeting the C- and N-terminals of COL6A3, found that the causal effect was driven by the C-terminal, cleaved into the adipokine endotrophin.
3/10
January 24, 2025 at 1:41 PM
Then, we evaluated the causal effect of BMI-driven proteins on CAD, stroke, and type 2 diabetes. We further filtered with colocalization and mediation analyses, identifying five protein mediators with six associations.
2/10
January 24, 2025 at 1:41 PM
We used a two-step proteome-wide MR approach to identify a protein mediator for this relationship.
First, we used MR to evaluate the causal effect of obesity on plasma proteins to identify “BMI-driven proteins.”
1/10
January 24, 2025 at 1:41 PM
Key highlights:
• ↓Body fat → ↓endotrophin → ↓CAD risk
• Two-step proteome-wide MR identifies proteins linking risk factors to disease
• Domain-aware MR shows only C-terminal COL6A3, cleaved into endotrophin, is causal for CAD
• Supports Endotrophin as a potential CAD target
January 24, 2025 at 1:41 PM
📣Excited to share our paper in Nature Genetics!
rdcu.be/d7mo0

Do plasma proteins mediate obesity’s effect on CAD risk?

→Using two-step proteome-wide MR, domain-aware MR, epigenomics & scRNA-seq, we prioritized endotrophin, cleaved from COL6A3, as a mediator & potential therapeutic target.
A 🧵↓
January 24, 2025 at 1:41 PM
Honored to work with these amazing people. Thank you so much!
10/10
January 24, 2025 at 1:14 PM
Overall, our study provides strong human genetic evidence supporting C-terminal COL6A3-derived endotrophin as a mediator of obesity’s effect on coronary artery disease and highlights it as a potential therapeutic target.
9/10
January 24, 2025 at 1:14 PM
Nice—but drug development takes time. In the meantime, can we do something? The answer is yes.
Using multivariable MR, we showed that reducing body fat (and for some proteins, increasing muscle) can lower protein mediator levels and cardiometabolic risk!
7/10
January 24, 2025 at 1:14 PM
We also found that the cis-pQTL of C-COL6A3 (rs11677932) overlaps with active enhancer regions in adipose tissue and the coronary artery. rs11677932 also disrupts DNA binding of a transcription factor, consistent with its association with ↓expression and plasma levels of COL6A3.
6/10
January 24, 2025 at 1:14 PM
In human tissue data from GTEx, COL6A3 was expressed in multiple tissues. We took a closer look at relevant tissues (adipose tissue and coronary artery) with single-cell RNA sequencing. COL6A3 was highly expressed in tissues involved in metabolic dysfunction and fibrosis.
5/10
January 24, 2025 at 1:14 PM